Seventy-four patients aged 26-63 yr who had suffered cervicogenic headache for 3 months-21 yr were treated with puked radiofrequency applied to C2 dorsal root ganglion, which is located in the middle of the posterior side of lateral atlantoaxial joint. A trochar was introduced percutaneously under the guidance of X-ray aiming at the target point. As it was inserted through the deep fascia, the stylet was withdrawn and a 10 cm long 22 gauge curved blunt electrode was inserted into the trochar and advanced until the patients felt radiating pain from the point of puncture to occiput. Lateral radiograph was obtained to verify the placement of electrode. The tip of the electrode was usually located in front of spinal canal at the atlantoaxial joint level. Sensory stimulation was performed with 50 Hz and 0.1-0.5 V and the patients could feel radiating pain at occiput. Motor stimulation was performed with 2 Hz and 0.4-1.0 V and regular pulsation of the patient's muscle of occiput could occur. Pulsed radiofrequency was applied at 42 ℃7 for 240 s and was performed twice on each side. VAS scores and disturbances of daily activity, mood and sleep were recorded before operation and at 1 week and 1, 3, 6, 12 and 18 months after pulsed radiofrequency treatment. Complications and recurrence within 12 and 18 months were recorded. Follow-up was lost in 22 patients. VAS scores and disturbances of daily activity, mood and sleep significantly decreased after procedure. All of the patients responded without complications like infection, spinal cord and vertebral artery injury. Some patients had transient occipital neuralgia which was usually relieved within 24 h. The recurrence rate in 12 and 18 months after operation was 19% and 31% respectively.

From November 2003 to May 2010, intrathecal drug delivery system (IDDS) was implanted in 18 patients with chronic intractable pain. Analgesia was provided with morphine. Thirteen patients suffered from late stage cancer and 5 from diseases other than cancer. VAS score was used to measure intensity of pain in all 18patients. QLQ-C30 score was used to evaluate quality of life in cancer patients. The patients were followed up for 3-62 months in 5 non-cancer patients. All 13 cancer patients died at 57 days-10 months after operation. VAS scores were significantly decreased and QLQ-C30 scores increased by intrathecal administration of morphine. Side effects developed in all patients to some extent including nausea, vomiting, constipation, urinary retention, pruritus and over-sedation and vanished in a week. Intrathecal catheter was cut while being pulled out of the needle in 1 patient. Two patients developed low intracranial pressure after operation. Cerebrospinal fluid leakage occurred in 1 patient. One patient developed neuropathic pain in the posterolateral side of right leg.

Objective To evaluate the effect of intrathecal oxytocin on neuropathic pain in rats.Methods Thirty-six SPF male Sprague-Dawley rats,aged 4-6 weeks,weighing 100-150 g,were divided into 4 groups (n =9 each) using a random number table:control group (group C),neuropathic pain group (group NP),neuropathic pain plus normal saline group (group NPN) and neuropathic pain plus oxytocin group (group NPO).The neuropathic pain model was made by partial sciatic nerve injury in NP,NPN and NPO groups.In group NPO,oxytocin l0 μl (0.1 μg) was intrathecally injected on the day of establishing the model and 1 and 2 days after establishing the model,and then normal saline 10 μl was given for tube sealing at 9 a.m.and 4 p.m.every day.In group NPN,normal saline 20 μl was given for tube sealing at the corresponding time points.The mechanical paw withdrawal threshold (MWT) and thermal paw withdrawal latency (TWL) were measured at 1 day before establishing the model and 1,2,3,4,5,6 and 7 days after establishing the model.The expression of the astrocyte marker glial fibrillary acidic protein (GFAP) and a specific marker of microglia ionized calcium-binding adaptor molecule 1 (IBa-1) was detected by Western blot at 1 day before establishing the model and 3 and 7 days after establishing the model.Results Compared with group C,the MWT was significantly decreased,TWL was shortened and the expression of GFAP and IBa-1 was up-regulated at each time point in NP and NPN groups (P＜0.05).There was no significant difference in MWT,TWL GFAP and IBa-1 at each time point between group NPN and group NP (P＞0.05).Compared with NP group,MWT was significantly increased at 2-4 days after establishing the model,TWL was prolonged at 1-4 days after establishing the model,the expression of IBa-1 was down-regulated on 3 days after establishing the model,and the expression of GFAP was downregulated on 3 and 7 days after establishing the model in group NPO (P＜0.05).Conclusion Oxytocin can reduce neuropathic pain,and the mechanism may be related to inhibiting activation of glial cells in the spinal cord of rats.